CN1341237A - Access device for semiconductor memory card, computer readable recording medium, initialization method, and semiconductor memory card - Google Patents

Abstract

In a semiconductor memory card, a predetermined number of erasable blocks located at the beginning of a volume area contain volume management information. The user area next to the volume management information includes a plurality of clusters. For the section from the master boot record and partition table sector to the partition boot area, the data length NOM is determined such that the plurality of clusters of the user area are arranged not to cross the boundary of the erasable block. Since the cluster boundary and the erasable block boundary coincide in the user area, it is not necessary to perform a wasteful process of erasing two erasable blocks in order to rewrite one cluster.

Description

Access device for semiconductor memory card, computer readable recording medium, initialization method, and semiconductor memory card

technical field

The invention relates to an access device for accessing a semiconductor memory card containing EEPROM (Electrically Erasable Programmable Read-Only Memory) nonvolatile memory, and a computer-readable recording medium for recording the initialization program of the semiconductor memory card , an initialization method and a semiconductor memory card, in particular, an improved method capable of improving data rewriting efficiency in the non-volatile memory.

technical background

Semiconductor memory cards have the advantage of being small in size and light in weight, and are developing smoothly in order to consolidate their position as an optional recording medium in various technical fields. A semiconductor memory card has a built-in nonvolatile memory called EEPROM, which is accessed by a connected device, so this semiconductor memory card can be used as a recording medium. And it is possible to directly write data into blank EEPROM sectors in the same way as magnetic disks or optical discs. However, when there are data stored in the EEPROM sector, these data must be deleted to restore these sectors to a blank state, so that new data can be written into the sector. In the EEPROM called NAND (NAND) used in many semiconductor memory cards, the operation of restoring a sector to a blank state must be performed simultaneously in 32 sectors (in this non-volatile memory, one A group of 32 sectors is called an erasable block). Therefore, the semiconductor memory card includes a special internal control circuit to implement memory management by using an erasable block as an access unit. The state control of the erasable block and the data reading and writing to the erasable block are all managed by the control circuit.

This means that the semiconductor memory card has a unique hardware structure (physical layer) that is completely different from that used in magnetic or optical disks. However, the layer model shown in Figure 1A consists of a physical layer, a file system layer, and an application layer in the same way as is used in magnetic or optical disks. Figure 1B shows the detailed configuration of the physical layer. In this figure, the physical layer includes a volume area in which a plurality of erasable blocks are arranged. Each erasable block consists of 32 sectors, and the data length is 16KB. The configuration of the file system layer shown in FIG. 1C is what is commonly called a FAT (File Allocation Table) file system. In the FAT file system, the management of the volume area is performed in units called clusters. Volume management information is set at the beginning of the volume area, and next to the volume management information is a user area for recording user data. The volume management information includes: master boot record, partition table, partition boot area, double file allocation table (FAT) and root directory entry. The double FAT represents a link between a plurality of clusters contained in a volume area. Provision of such a file system layer enables data to be stored in the application layer in a hierarchical structure formed of directories and files. By setting the layer model, the access device can access the semiconductor memory card using the same program as for accessing recording media such as magnetic disks or optical discs, without paying attention to the difference in the physical layer.

However, when storing data in the volume area of the file system, the user has multiple possibilities to determine the data length of the volume area. When the size of the volume area changes according to the user's needs, the number of clusters contained in the volume area will increase or decrease accordingly. If the number of clusters increases or decreases, the FAT composed of entries corresponding to these clusters will correspondingly increase or decrease, and so will the size of the volume management information including the FAT. If the size of the volume management information increases or decreases, then the start address of the user area of the volume management information will also change. The user area start address changes according to the size of the volume area. Therefore, the start address of each cluster also changes according to the size of the user area.

If the start address of each cluster changes according to the size of the user area, the cluster may straddle the boundary between two erasable blocks, and the end of the volume management information may be located at the beginning of the user area A cluster at is set in the same erasable block. FIG. 1D shows the volume area configuration when the end part of the volume management information and the cluster located at the beginning of the user area are in the same erasable block. If clusters are set as shown in the figure, and the user wishes to modify data stored in a specific cluster, the two erasable blocks in which the cluster is set must be read and then restored to a blank state. However, nonvolatile memory consists of a memory element in which a floating gate is buried in an insulating layer. This storage element can only be erased tens of thousands of times, so if it is a frequent occurrence that two erasable blocks have to be erased in order to modify a cluster, the lifetime of the non-volatile memory will be significantly shortened.

Generally speaking, if the erasure of the write target cluster has been completed, when 32 sectors are managed as a cluster, information can be written to one cluster in 32×200 μs (200 μs is the time required to write each sector) time) to complete. However, if the write target must be erased before data can be written, a 2ms erase period must be added. If the cluster straddles the boundary between two erasable blocks, both blocks need to be erased, and it takes 4ms to erase the write target. Therefore, the time taken to write data increases significantly.

Contents of the invention

The purpose of the present invention is to provide an access device to shorten the processing time required to modify an erasable block, and to construct a storage data format in a semiconductor memory card, so that the life of the non-volatile memory is longer.

A type of semiconductor memory card called a Secure Digital (SD) memory card has made great progress in solving the above-mentioned problems of shortening the processing time and prolonging the life of fixed memory cards for the following reasons. In the SD memory card, an area called a protected area is provided, and general users cannot use this area. This protected area is used to store secret information such as encryption keys used to encrypt data, billing information used to notify users to pay when copying copyrighted materials, and the like. The amount of data that needs to be kept secret varies according to the type of application used, so the size of the protected area must also vary according to the type of application. If the size of the protected area changes, the configuration of the volume area also changes according to the type of application. If the configuration of volume areas is changed in this way, some clusters in this configuration will often straddle the boundaries between erasable blocks, so this is particularly desirable. In order to achieve the above object, the semiconductor memory card access device can have the following structure: manage the data on each group of 2 ^j (j is 0 or a positive integer) sectors as a cluster, and manage one or more clusters It is managed as a file, and the access device implements file access on a semiconductor memory card containing a storage area composed of multiple sectors. Here, each group of 2 ⁱ consecutive sectors in the storage area constitutes a block (i is 0 or a positive integer), and a block is the smallest unit capable of erasing data. The access device includes a calculation unit, a reservation unit and a recording unit. The calculation unit calculates the size of the volume management information according to the number of clusters to be managed in the storage area. Here, the volume management information includes a file allocation table for indicating, for each file, links between clusters corresponding to the file. The reservation unit reserves a first area for recording volume management information and a second area for recording user data. The first area has a data length larger than the calculated volume management information and is composed of m× ^2j (m is a positive integer) sectors, and the second area is composed of sectors following the first area. The recording unit records volume management information in the first area, records user data in the second area, and manages these volume management information and user data as a plurality of clusters. In the access device, an area is reserved in the volume area containing m (m is a positive integer) clusters for recording volume management information, so that there will be no stored clusters spanning two Possibility of rubbing blocks. A cluster boundary can coincide with a boundary of an erasable block, and a boundary of the volume management information can coincide with a boundary of an erasable block. Therefore, when rewriting or rewriting a cluster, only one erasable block needs to be erased, thus reducing the number of erasing times necessary for an erasable block. If the erasing times of the erasable blocks are reduced, the semiconductor memory card can complete data writing in a short time, and the life of the nonvolatile memory itself can be increased.

Here, in addition to the file allocation table, the volume management information may also include: master boot record, partition table, partition boot area information, and root directory entries. In addition, the recording unit records the master boot record and partition table in the first sector of the first area, skips a predetermined number of sectors and records the partition boot area information, File allocation table and root directory entry. Thus, the end of the first area may coincide with the end of the root directory entry. The number of sectors between the master boot record representing the start of the drive and the partition boot area representing the start of the partition can be adjusted, so the volume management information can be limited to the first area consisting of m sectors, and Compatible with devices using the saved FAT file system.

Here, the calculation unit may calculate the sum SUM by adding the number of sectors used to record the partition boot area information, file allocation table and root directory entry. The reserving unit reserves the first area by calculating the value of m according to {Formula 1} NOM+SUM=2 ^j +m. Among them, NOM is the number of sectors. The recording unit calculates the predetermined number of sectors by subtracting 1 from the number of sectors NOM. Even if the size of the file allocation table changes, a first area will still be reserved, which is larger than the size of the volume management information and is an integer multiple of the erasable block size. Therefore, no matter how the size of the file allocation table is calculated, the minimum size necessary for the first area can always be reserved.

Here, the recording unit may set a predetermined number of sectors in the partition table recording the volume management information. In this structure, even if the size of the first area changes and causes the start address of the second area to change, a certain number of sectors will still be set in the partition table. The area number NOM is obtained by subtracting 1, so that the access device can accurately access the user area by referring to the partition table.

Brief description of the drawings

Figure 1A shows a configuration model consisting of a physical layer, a file system layer compliant with ISO/IEC9293 and an application layer;

Figure 1B and 1C show the format of the physical layer and the file system layer;

Figure 1D shows a volume area configuration when the end part of the volume management information and a cluster at the beginning of the user area are arranged in the same erasable block;

Fig. 2A is the external view of semiconductor memory card;

2B and 2C are external views of the access device;

Figure 3A shows the internal structure of the semiconductor memory card and the access device;

Figure 3B shows a layer model of software used in the access device;

Figure 4 shows a data storage format used by the non-volatile memory 1, and complies with the ISO/IEC9293 standard;

Figure 5 shows the structure of the partition control area, system area and user area included in the volume area;

Fig. 6A schematically represents the structure of master boot record and partition table sector;

Figure 6B shows the structure of the partition boot area;

Figure 7A shows the structure of double FAT;

Figure 7B shows the structure of the root directory entry;

Figure 7C shows the structure of the user area;

Figure 8 shows an embodiment of the file storage method;

Fig. 9 shows that when file AOB001.SA1 is stored in a plurality of clusters, the embodiment that described root directory entry and FAT are set;

Figure 10A shows the relationship between erasable blocks and clusters;

Figure 10B shows multiple clusters when n=1;

Figure 10C shows multiple clusters when n=16;

FIG. 11A assumes a state when m clusters among s erasable blocks are allocated for recording volume management information;

Figure 11B shows the number of clusters allocated for recording volume management information when n=1;

Figure 11C shows the number of clusters allocated for recording volume management information when n=16;

FIG. 12 is a diagram obtained by plotting the partition control area, system area, and cluster by the sizes calculated using Equations 11 and 12;

Fig. 13 is a flowchart representing the initialization process of the volume area;

Fig. 14 shows the structures of the access device and the semiconductor memory card in the second embodiment;

Figure 15 shows the internal structure of the security processing unit 11;

Figure 16 shows the detailed structure of volume areas provided with protected areas;

Figure 17 shows an exemplary configuration of the user data area;

Figure 18 shows an exemplary configuration of the protection zone;

Fig. 19 shows the internal structure of the access control unit 2 in the third embodiment;

20A ~ 20D show the processing sequence when the erasable block is rewritten;

Fig. 21 shows the internal structure of the file system operating unit 9 in the third embodiment;

FIG. 22 is a flowchart showing the detailed processing procedure performed by the file system operation unit 9 in the third embodiment;

Figures 23A and 23B show the erasing process performed in an erasable block in the third embodiment;

Fig. 24 shows the internal structure of the access control unit 2 in the fourth embodiment;

FIG. 25 is a flowchart showing the detailed processing procedure executed by the file system operation unit 9 in the fourth embodiment;

Figures 26A and 26B show the erasing process performed in an erasable block in the fourth embodiment;

Figures 27A and 27B show the causal relationship between memory segments and auxiliary operations when a command is issued;

Fig. 28 is a flowchart showing a detailed processing sequence of memory fragment erasure processing executed on a logical address;

Fig. 29A ~ 29D shows the hypothetical model corresponding to each variable s, t, u, v and y in the flowchart of Fig. 28;

30A ~ 30C show how to eliminate memory segments in the fifth embodiment;

Figure 31 shows that a new extended attribute is defined by using the implementation purpose extended attribute in UDF;

Figure 32 shows the internal structure of the semiconductor memory card containing the backup area;

Fig. 33 shows the internal structure of the semiconductor memory card in the sixth embodiment; and

34A to 34C show the contents of processing executed by the file system operation unit 9 in the seventh embodiment.

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiments of the semiconductor memory card and the system including the semiconductor memory card and the access device will be described below with reference to the accompanying drawings.

2A is an external view of the semiconductor memory card 100, and FIGS. 2B and 2C are external views of the access device. FIG. 3A shows the internal structure of the semiconductor memory card 100 and the access device 200 .

The semiconductor memory card has an external structure as shown in FIG. 2A , with a length of 32.0mm, a width of 24.0mm, and a thickness of 2.1mm. It is about the size of a postage stamp and small enough to be placed on a user's fingertip. The semiconductor memory card 100 has nine connectors for connecting an access device 200, and has a write protection switch 101 on one side which can be set by the user to allow or prohibit rewriting of recorded data. As shown in the lower part of FIG. 3A, the semiconductor memory card 100 includes a nonvolatile memory 1 made of NAND EEPROM, an access control unit 2 and a working memory 3. The access control unit 2 writes data into the non-volatile memory 1 , reads data from the non-volatile memory 1 , and erases data according to instructions issued by the access device 200 . When the data read from the non-volatile memory is to be rewritten or written back to the non-volatile memory, the working memory 3 is used for temporarily storing these data.

Next, the access device 200 is described, for example, a home audio system shown in FIG. 2B or an information processing device such as a personal computer shown in FIG. 2C. This access device 200 includes: a card connector 4, an auxiliary memory 5, a CPU 6 and a main memory 7. The card connector 4 is used for connecting the semiconductor memory card 100 . The auxiliary memory 5 stores software for accessing the semiconductor memory card 100 . The CPU 6 performs overall control of the access device 200. The main memory 7 is used to temporarily store the FAT and root directory entries when accessing the semiconductor memory card 100 . FIG. 3B shows a configuration model of software used in the access device 200 . In this figure, the access device software includes: an application program 8 , a file system operating unit 9 and a device driver 10 . The application program 8 executes predetermined processing such as audio and video reproduction for the access device 200 . The file system operation unit 9 executes read, write, erase and modify (rewrite) operations of files in the file system according to instructions from the application program 8 . The device driver 10 performs operations in the file system by issuing read and write commands to the semiconductor memory card 100 .

One embodiment of the data storage format of the nonvolatile memory 1 is explained below. This nonvolatile memory 1 performs data storage in the format shown in FIG. 4 . In this figure, the entire non-volatile memory 1 is called the volume area. The volume area is divided into multiple clusters, and includes a partition control area and a partition (also known as a regular area). The partition is divided into a system area and a user area, as shown on the right hand side of Figure 4.

The starting address of the user area is directly followed by the system area. However, the size of the double FAT in the system area varies according to the size of the user area, so, as described in the technical background section, the start address of the user area will also change accordingly. Each sector in the volume is indicated by a physical address relative to the beginning of the volume.

Each area contained in this volume area is described in turn below. Fig. 5 shows the structure of the partition control area, system area and user area included in the volume area.

The partition control area includes master boot record and partition table sectors, and reserved information 1/2 and 2/2. Fig. 6A shows the detailed structure of the master boot record and partition table sectors. In this figure, the contents of the MBR and partition table sectors are listed hierarchically between a pair of arrows ky1. The master boot record and partition table sector includes a master boot record, four partition tables 1, 2, 3 and 4, and a flag word.

The master boot record is a symbolic indication of the access device 200, and its subsequent area is a physical medium (a piece of physical medium). In FIG. 6A , the volume area has only one master boot record, so the access device 200 recognizes the volume area as a physical medium. Usually, if there are two master boot records in the volume area, the access device 200 will recognize them as two pieces of physical media.

The partition table is a table recording partition information. As shown by arrow ky2, when the semiconductor memory card 100 is used as a boot driver, the partition table includes the following fields: 'boot flag', 'start end', 'start sector/start cylinder', ' System ID', 'Ending End', 'Ending Sector/Ending Cylinder', 'Relative Sector' and 'Total Sectors'. The 'Boot Flags' field is set to '0x80'. The 'Start' field defines the start of a partition. The 'Start Sector/Start Cylinder' field defines the start sector and start cylinder of the partition. 'System ID' defines a file system type, and this field is set to '01' when the size of the partition is less than 32680 bytes, and is set to '04' when the size of the partition is less than 65536 bytes. The 'End Sector/End Cylinder' field defines the end sector and end cylinder of the partition. The 'Relative Sector' field defines the number of sectors that exist before the start sector of the partition. The 'Total Sectors' field defines the number of sectors in this partition.

The partitions in the semiconductor memory card 100 will be described below. This partition consists of the system area, followed by the user area, but this description first addresses the user area with reference to Figure 7C.

The user area stores files in units not smaller than one cluster. The dotted arrow ff2 in FIG. 7C indicates several clusters 002, 003, 004, 005, . . . included in the user area. The numbers 002, 003, 004, 005, 006, 007, 008... used in Fig. 7C are three-digit hexadecimal cluster numbers used only to identify each cluster. Since the minimum unit in which access can be performed is a cluster, a cluster number is used to indicate a storage location in the user area.

The system area includes partition boot area, dual file allocation table and root directory entry. The partition boot area, dual file allocation table FAT and root directory entries will be described in sequence with reference to FIGS. 6B , 7A and 7B.

An extended FDC (Floppy Disk Controller) descriptor shown in FIG. 6B containing the information field is set in the partition boot area. The extended FDC descriptor includes the following fields: 'transfer instruction', 'establish system identifier', 'sector size', 'number of sectors per cluster', 'reserved sector count', 'FAT number' (included in FAT number in the double FAT), 'root directory entry number' (data length of the root directory entry), 'total sector', 'medium identifier', 'sector number of each FAT', 'Number of sectors per track', 'Number of magnetic surfaces', 'Number of hidden sectors', 'Number of total sectors' (the total number of sectors in the system area and user area), 'Number of physical disks', 'Expansion Boot record flag', 'volume ID number', 'volume label', 'file system type' and 'flag word'.

The dual FAT is composed of two FATs conforming to the ISO/IEC (International Organization for Standardization/International Electrotechnical Commission) 9293 standard. Each FAT includes multiple FAT entries, and each FAT entry is linked to a cluster. The FAT entry indicates whether the corresponding cluster is used: if the cluster is not used, the entry is set to '0'; if the cluster is in use, it is set to the corresponding cluster number. The cluster number indicates a link to the next cluster that needs to be read immediately after this cluster. The dotted arrow ff1 drawn in FIG. 7A indicates a plurality of FAT entries 002, 003, 004, 005, . . . contained in the FAT. These numerical values attached to each FAT entry indicate the cluster number of the corresponding cluster.

The root directory entry includes a plurality of file entries corresponding to a plurality of files existing in the boot directory. Each file registration item includes: the 'file name' of the existing file, the 'file identifier', the 'first cluster number of the file' stored at the beginning of the file, the 'file attribute', and the 'storage number' indicating when the file is stored time' and 'storage date', and 'file length'.

The file storage method will be explained below by explaining how to store the file name 'AOB001.SA1' in the root directory with reference to FIG. 8 . Since the minimum unit that can access the user area is a cluster, the file 'AOB001.SA1' needs to be stored in the data area of sectors not smaller than a cluster. Therefore, the file 'AOB001.SA1' is first divided into clusters and then stored. In Figure 8, the file 'AOB001.SA1' is divided into five sectors according to the cluster size, and the divided sectors are stored in clusters numbered 003, 004, 005, 00A and 00C.

The embodiment shown in Fig. 9 is how to set and store the root directory entry and FAT when the file 'AOB001.SA1' which has been divided into multiple sectors is stored. In this figure, the beginning of the file 'AOB001.SA1' is stored in cluster 003, so the cluster number 003 is written in the 'first cluster number of the file' of the root directory entry to indicate that the cluster stores is the first snippet of the file. Subsequent segments of the file 'AOB001.SA1' are stored in clusters 004 and 005. Therefore, when the entry 003 (004) of the FAT corresponds to the cluster 003 storing the first fragment of the file 'AOB001.SA1', the entry indicates that the cluster 004 is the cluster storing the next fragment of the file 'AOB001.SA1'. Similarly, when the entries 004 (005) and 005 (00A) of the FAT correspond to the clusters 004 and 005 storing the next segment of the file 'AOB001.SA1', these entries respectively represent that the clusters 005 and 00A are storing the Cluster for the next segment of the file 'AOB001.SA1'. As shown by the arrows fk1, fk2, fk3, fk4, fk5... in Figure 9, by reading those clusters whose cluster numbers are sequentially written into the FAT entries, all sectors obtained by dividing the file 'AOB001.SA1' area can be read. As mentioned above, access to the user area of the semiconductor memory card 100 is performed in units of clusters, and each cluster is associated with a FAT entry. It should be pointed out that: the FAT entry corresponding to the cluster (cluster 00C in the embodiment of FIG. 9 ) storing the end fragment of the AOB file is set to the cluster number 'FFF' to indicate that the corresponding cluster stores the end fragment of the file.

The above explanations give an overview of the file system in the fixed storage 1 of the present invention. The following description will focus on the principle of the implementation, explaining how to adjust the cluster boundary and the boundary of the erasable block, that is, how to make the boundary between the system area and the user area consistent with the boundary between two erasable blocks. Two variants of this embodiment aim at achieving this boundary alignment. The first improvement is to set the cluster size to 1/n of the erasable block size (n is 1, 2, 4, 8, 16 or 32). Figure 10A shows the relationship between erasable blocks and clusters. Here, a cluster is defined as 1/n of an erasable block size, namely: 1/n of 16KB (1/n of 32 sectors). FIG. 10B shows multiple clusters when n=1, and FIG. 10C shows multiple clusters when n=16.

The second improvement solution is to designate an area in the volume area, the size of which is m times that of a cluster, for recording the volume management information. FIG. 11A assumes a state when m clusters of s clusters among s erasable blocks are allocated for recording volume management information. If m clusters are allocated for recording volume management information, the volume management information will occupy m/n areas of these s erasable blocks, and the remaining (s·n-m)/n areas will be classified as user areas.

By setting the size of the volume management information to be m times of the cluster size, the volume management information and s·n-m clusters can be compressed so that a cluster does not straddle the boundary between two erasable blocks.

FIG. 11B shows the number of clusters allocated for recording volume management information when n=1, and FIG. 11C shows the number of clusters allocated for recording volume management information when n=16. These figures clearly show that by setting the volume management information, multiple clusters are exactly matched with multiple erasable blocks so that no cluster crosses the boundary between erasable blocks. The detailed setting of the volume area whose size has been adjusted according to this method is shown in FIG. 5 . In this figure, the size of the partition control area is NOM, the size of the partition boot area is RSC, the size of the double FAT is 1×2, the size of the root directory entry is RDE, and the total sector size is TS. The number of sectors contained in a cluster is SC.

In FIG. 5, the size "size 1" of the FAT in the dual FAT is determined based on the total sector size TS. In particular, this value is calculated using Equation 11 below. Formula 11

For FAT12: Size 1(((((TS－(RSC+RDE))/SC)+2)×12/8)+511)/512

For FAT16: size 1(((((TS－(RSC+RDE))/SC)+2)×16/8)+511)/512

Here, FAT12 represents a file system that allocates 12 bits to each FAT entry, and FAT16 represents a file system that allocates 16 bits to each FAT entry.

In Formula 11, (TS-(RSD+RDE)) is the number of clusters required for recording in the user area. And this number adds indicator number 2, its result is multiplied with the byte length (12 or 16) of this FAT entry again, divides by 8 then and just obtains the byte length of this FAT. Add an offset value of 511 bytes to the calculation result, and divide the result by the byte length 512 of one sector to calculate the number of sectors required to store one FAT. If the data length of the FAT is 1 byte, dividing the FAT data by 521 will result in the sector number of the FAT being calculated as 0. However, Equation 11 plus an offset value of 511 will ensure that the calculated size 1 has at least one sector.

In Figure 5, an important point to point out is the size of the partition control area called NOM. Set the value of NOM to 'SC+α' to ensure that the partition control area and system area belong to different clusters. The reason for setting NOM in this way is to prevent the impact on the partition control area when a mistake occurs in modifying the cluster in the system area. In other words, if the master boot record and partition table sectors in the partition are damaged by such modification errors, the semiconductor memory card 100 will no longer be recognized by the access device 200 as an authorized recording medium. In order to avoid this unfavorable situation, the value of NOM is set to SC+α.

The setting of SC+α value is carried out as follows. Divide (RSC + RDE + size 1 × 2) by SC to calculate the α value, so that the sum of RSC, size 1 × 2 and RDE is an integer multiple of SC, and subtract the remainder of this calculation from SC. If these factors are taken into account, NOM is calculated according to Equation 12. Formula 12

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

If α is defined in this way, the partition control area and user area will be formed so as to fit a plurality of erasable blocks exactly, and the boundary between the system area and user area will coincide with the boundary of an erasable block. If this boundary alignment can be achieved, then the boundaries of all subsequent clusters will coincide with the boundaries of the erasable block.

Here, the embodiment illustrates how to calculate the NOM and size 1 when the SC is 32 sectors (16KB), the RDE is 32 sectors, the RSC is 1 sector, and the TS is 65600 sectors. If the type of the double FAT is FAT12, the size 1 is calculated as follows using Equation 11.

Size 1 = (((((TS－(RSC+RDE))/SC)+2)×12/8)+511)/512

＝(((((65600－(1＋32))/SC)＋2)×12/8)＋511)/512

= 7 sectors

In addition, the NOM is calculated using Equation 12 as follows.

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

＝(32－(1+32+7×2)%32)+32

= 47 sectors

Fig. 12 lists the partition control area, system area, and cluster according to the calculated sizes. The master boot record and partition table sectors, and reserved information 1/2 are arranged in sectors PSN000~PSN031, and the reserved information 2/2, partition boot area and double FAT are arranged in sectors PSN032~PSN063. The root directory entries are arranged in sectors PSN064~PSN095. In this volume area, 32 sectors constitute an erasable block, therefore, the master boot record, partition table and reserved information 1/2 are arranged in the first erasable block 001, and the reserved information 2/ 2. The partition boot area and double FAT are arranged in the second erasable block 002, and the entry of the root directory is arranged in the third erasable block 003. The root directory entries are stored in erasable blocks of the same size so that the boundary between user area and system area matches the boundary between erasable blocks.

When the file system operation unit 9 executes initialization on the non-volatile memory 1, the configuration of the above-mentioned volume area will be realized. A procedure for executing this initialization processing will be explained with reference to the flowchart of FIG. 13 .

In step S1, according to the size (TS) of the area to be formatted, and the operating system used by the semiconductor memory card 100, the access device 200 and the total storage capacity of instructions from the user, it is determined by the file system operating unit 9 The size of the cluster.

Once the cluster size is determined, at step S2 the file system operation unit 9 uses the cluster size SC and the total sector size TS to determine which file system should be used: FAT12 or FAT16. Once one of the file systems FAT12 and FAT16 is determined, the file system operation unit 9 determines the length RDE of the root directory entry in step S3 (RDE is determined as 32 sectors in these embodiments), and then, in step S3 In S4, determine the length RSC of the boot area of the partition (in these embodiments, RSC is defined as 1 sector). Once the RDE and RSC are obtained, the file system operation unit 9 will use formula 11 to calculate the data length of the FAT in step S5. Then, in step S6, the file system operation unit 9 calculates the NOM using formula 12, so that the master boot record and the partition boot area are located in different clusters.

The processing performed in the above-mentioned steps S5 to S7 is actually exactly the same as the processing performed in the aforementioned formulas 11 and 12. However, in this flowchart, the processing executed in steps S7 to S9 will be described first. In step S7, the file system operation unit 9 calculates the number of clusters CN in the user area using formula 13. Formula 13

CN=(TS-(RSC+RDE+Size 1×2+NOM))/SC

In step S8, the file system operation unit 9 calculates the data length of the FAT using the following formula 14. Formula 14

For FAT12: Size 2 = (((CN+2)×12/8+511)/512

For FAT16: Size 2 = (((CN+2)×16/8+511)/512

In step S9, the file system operation unit 9 compares the size 1 calculated in step S5 with the size 2 calculated in step S8, and ends the process if the two values are equal. If the two values are not equal, the file system operation unit 9 replaces size 2 with size 1 at step S10, goes to step S6, and calculates the NOM. Since dimension 1 is now a different value due to substitution, a different NOM can be calculated by going to step S6. Then, the file system operation unit 9 recalculates size 2 in step S8 according to the newly calculated NOM, and if size 1 and size 2 are equal, then step S9 is "Yes" and the process ends.

'ab The NOM calculated by the above processing is put into the 'relative sector' field in the partition table to indicate the number of sectors that exist before the starting sector of the partition, and TS is put into the 'relative sector' field in the partition table 'Total Sectors' field.

The SC is put into the field of 'number of sectors per cluster' in the boot area of the partition to indicate the number of sectors contained in each cluster. In addition, multiply the RDE representing the number of sectors by the sector length of 512 bytes, and then divide by 32 to obtain the number of file entries, and then put the number into the 'root directory entry number' in the boot area of the partition ' field. Put size 1 into the field of 'the number of sectors per FAT' in the boot area of the partition, indicating the number of sectors in each FAT. When the access device 200 determines the positions of the double FAT, the root directory entry and the user area, it will refer to these values placed in the partition table and partition boot area.

This is an explanation of the flowchart in FIG. 13 . The following example calculations will illustrate how the partition and system sizes are calculated when the TS is 65568.

Since TS is 65568, in step S1, the file system operation unit 9 determines the cluster size as 32 sectors. Once the size of the cluster is determined to be 32 sectors, in step S2, the file system operation unit 9 determines that the FAT12 file system should be used according to the size SC of the cluster and the size TS of the total sectors. In these embodiments, RDE and RSC are set to be 32 sectors and 1 sector respectively, so the calculations of steps S3 and S4 are not performed. In step S5, the file operating system uses formulas 11 and 12 to calculate the data length of the FAT. Here, since the FAT is FAT12, the following calculation is performed.

Size 1 = (((((TS－(RSC+RDE))/SC)+2)×12/8)+511)/512

＝(((((65568－(1＋32))/SC)＋2)×12/8)＋511)/512

= 7 sectors

Once size 1 is calculated, at step S6, the file system operation unit 9 calculates NOM using Equation 12 so that the master boot record and partition boot area are located in different clusters.

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

＝(32－(1+32+7×2)%32)+32

= 49 sectors

Once the NOM is calculated, the file system operation unit 9 calculates the number of clusters in the normal area using Equation 13 at step S7.

CN=(TS-(RSC+RDE+Size 1×2+NOM))/SC

＝(65568－(1＋32＋7×2＋49))/32

＝2046 sectors

In step S8, the file system operation unit 9 calculates the data length of the FAT using formula 14.

For FAT12: Size 2 = (((CN+2)×12/8)+511)/512

=(((2046+2)×12/8)+511)/512

= 6 sectors

In step S9, the file system operation unit 9 compares the size 1 calculated in step S5 with the size 2 calculated in step S8. Here, size 1 is 7 sectors, and size 2 is 6 sectors. Since the two values are not equal, the file system operation unit 9 goes to step S10, replaces size 2 with size 1, goes to step S6 and calculates the NOM.

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

＝(32－(1+32+6×2)%32)+32

= 51 sectors

Once NOM is calculated, at step S7, the file system operation unit 9 calculates the number of clusters in the normal area.

CN=(TS-(RSC+RDE+Size 1×2+NOM))/SC

(65568－(1＋32＋6×2＋49))/32

＝2046 sectors

In step S8, the data length of the FAT is calculated.

For FAT12: Size 2 = (((CN+2)×12/8)+511)/512

=(((2046+2)×12/8)+511)/512

=6

Following this calculation, at step S9, the file system operation unit 9 compares size 1 and size 2, and since these two values are equal, the processing of the flowchart ends.

As shown above, since NOM can be calculated in this embodiment, the size of the partition and system area is an integer multiple of the number of erasable blocks, thus ensuring that the storage of clusters will not span two erasable blocks. The boundary of the cluster is consistent with the boundary of the erasable block, and the boundary of the volume management information is consistent with the boundary of an erasable block. In this way, when a cluster is rewritten or rewritten, the number of erasable blocks that need to be erased is limited to one, thus reducing the number of times any erasable block needs to be erased. Therefore, the time required for writing data to the semiconductor memory card 100 is shortened, and the life of the nonvolatile memory 1 can be extended. second embodiment

The second embodiment proposes an arrangement in which the volume area is divided into an area (user data area) accessible to ordinary users and an area (protected area) storing confidential data.

The structures of the access device 200 and the semiconductor memory card 100 in the second embodiment are shown in FIG. 14 . In the internal structure shown in this figure, contrary to that shown in FIG. 3A, the nonvolatile memory 1 has provided a protected area, and the semiconductor memory card 100 and the access device 200 have provided security respectively. Processing units 11 and 12.

A secure processing unit 11 is provided which performs secure reading and writing of protected areas in said non-volatile memory 1, as explained below. As shown in FIG. 15 , the security processing unit 11 includes: a system area 13 , a hidden area 14 , an AKE processing unit 15 , a Ks decryption unit 16 and a Ks encryption unit 17 .

The system area 13 is a read-only area for storing a media key block (MKB) and a media-ID. MKB and Media-ID stored in this area cannot be rewritten. Assuming that the semiconductor memory card 100 is connected to the access device 200, the MKB and the medium-ID are read through the access device 200. If the access device 200 correctly performs specific calculations using the MKB, medium-ID and an internally stored device key Kd, the correct encryption key Kmu can be obtained.

The hidden area 14 stores the correct value of the encryption key Kmu, in other words, the encryption key Kmu should be obtained if the access device 200 performs a correct calculation using a valid device key Kd.

The AKE (authentication and key exchange) processing unit 15 performs mutual authentication between the access device 200 and the semiconductor memory card 100 using a challenge-response method, checks the authenticity of the relative device, and if the relative device is invalid, then Stop processing. However, if the relative device is valid, an encryption key (session key Ks) is shared between the access device 200 and the semiconductor memory card 100 . The verification performed by the access device 200 is divided into three stages: first, in the first inquiry stage, the access device 200 generates a random number, encrypts the random number with the encryption key Kmu, and sends the encrypted The random number is sent to the semiconductor memory card 100 as the query value A; secondly, in the first response phase, the semiconductor memory card 100 uses the internally stored encryption key Kmu to decrypt the query value A, and uses the decrypted value as a response The value B is passed to the access device 200; finally, in the first verification stage, the access device 200 uses its encryption key Kmu to decrypt the interrogation value A stored inside, and compares the decrypted value with the semiconductor storage The response value B transmitted from the card 100 is compared.

The verification performed by the semiconductor memory card 100 is also divided into three stages: first, in the second inquiry stage, the semiconductor memory card 100 generates a random number, encrypts the random number with the encryption key Kmu, and The encrypted random number is sent to the access device 200 as the query value C; secondly, in the second response stage, the access device 200 uses the internally stored encryption key Kmu to decrypt the query value C, and uses the decrypted value as The response value D is transmitted to the semiconductor memory card 100; finally, in the second verification stage, the semiconductor memory card 100 uses its encryption key Kmu to decrypt the internally stored query value C, and compares the decrypted value with the Compare with the response value D transmitted from the access device 200.

If the access device 200 performs mutual authentication using an invalid encryption key, the challenge value A and response value B of the first verification stage, and the challenge value C and response value D of the second verification stage will be judged to be invalid. matches the value, and mutual authentication will stop. However, if the authenticity of the relative device is confirmed, the AKE processing unit 15 will calculate the XOR value of the challenge value A and the challenge value C, and decrypt the XOR value by using the encryption key Kmu to obtain Obtain the dialogue keyword Ks.

When the access device 200 connected to the semiconductor memory card 100 outputs the encrypted data to be written into the protected area, the Ks decryption unit 16 decrypts the encrypted data that has been encrypted by the session key before output with the session key Ks. The data encrypted by the word Ks. Write the data obtained through the decryption into the protected area as original data.

Described Ks encryption unit 17 is received by the read data instruction that is sent by the access device 200 that is connected with described semiconductor memory card 100, encrypts the data that is stored in the protected area with described dialogue keyword Ks, then to access device 200 output these encrypted data. The reading and writing of data to the protected area is carried out after the Ks decryption unit 16 performs decryption and the Ks encryption unit 17 performs encryption, so only when the semiconductor memory card 100 and the connected access device 200 successfully The protected area can only be accessed when AKE processing is performed.

The volume format of the second embodiment is explained below with reference to the detailed view of the volume area including the protected area in FIG. 16 . In this figure, if the total size of the volume area is 'volume size', then the user area is a piece of 'volume size x (1-β)' extending from the beginning of the volume area, and will be followed by the user area The 'volume size x β' area is designated as the protected area. Among them, the minimum setting value of β is 0.01. For example, if the volume size is 64MB and β is 0.01, the protected area is set at 640KB. When the file system operation unit 9 executes the initialization process shown in FIG. 9, the formats of the user data area and the protected area can be realized.

The following is an exemplary calculation to calculate the sizes of the partition control area and the system area in the user data area when the total number of sectors TS in the user data area is 124160.

Since TS is 124160, in step S1, the file system operation unit 9 determines that the cluster size is 32 sectors. Once it is determined that the cluster size is 32 sectors, in step S2, the file system operation unit 9 determines that the FAT12 file system should be used with reference to the cluster size SC and the total sector size TS. In these implementations, the sizes of the RDE and RSC are set to 32 sectors and 1 sector respectively, so the calculations in steps S3 and S4 are not performed. In step 5, the file operating system calculates the data length of the FAT using formulas 11 and 12. Here, since the FAT is FAT12, the following calculation is performed.

Size 1 = (((((TS－(RSC+RDE))/SC)+2)×12/8)+511)/512

＝(((((124160－(1＋32))/32)＋2)×12/8)＋511)/512

= 12 sectors

Once size 1 is calculated, at step S6, the file system operation unit 9 calculates NOM using Equation 12 so that the master boot record and partition boot area are located in different clusters.

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

＝(32－(1+32+12×2)%32)+32

= 39 sectors

Once the NOM is calculated, the file system operation unit 9 calculates the number of clusters in the normal area using Equation 13 at step S7.

CN=(TS-(RSC+RDE+Size 1×2+NOM))/SC

＝(124160－(1+32+12×2+49))/32

＝3877 sectors

In step S8, the file system operation unit 9 recalculates the data length of the FAT using formula 14.

For FAT12: Size 2 = (((CN+2)×12/8)+511)/512

＝(((3877＋2×12/8)＋511)/512

= 12 sectors

In step S9, the file system operation unit 9 compares the size 1 calculated in step S5 with the size 2 calculated in step S8. Here, the calculated size 1 and size 2 are both 12 sectors, so it is judged that these two values are equal, and the process of this flowchart ends.

When the system area and partition control area are constituted by TS, size 1 and NOM calculated by using the above calculation formula, the configuration of the protection area is as shown in FIG. 17 . If you compare this figure with Figure 12, it can be seen that the size of the double FAT has been increased from 14 sectors to 24 sectors, and the size of reserved information 1/2 and 2/2 has been reduced from 17 sectors to 7 sectors In this way, the partition control area and the system area are arranged exactly in 3 erasable blocks.

The following description is an exemplary calculation. When the total number of sectors TS in the user data area is 1280, the sizes of the partition control area and the system area in the protection area are calculated.

Since TS is 1280, at step S1, the file system operation unit 9 determines that the size of the cluster is 2 sectors (1 KB of memory, and only 1/16 of the size of the user data area). Once it is determined that the cluster size is 2 sectors, in step S2, the file system operation unit 9 determines that the FAT12 file system should be used with reference to the cluster size SC and the total sector size TS. In these embodiments, the sizes of the RDE and RSC are set to 32 sectors and 1 sector respectively, so steps S3 and S4 are not performed. In step 5, the file operating system calculates the data length of the FAT using formulas 11 and 12. Here, since the FAT is FAT12, the following calculation is performed.

Size 1 = (((((TS－(RSC+RDE))/SC)+2)×12/8)+511)/512

＝(((((1280－(1＋32))/2)＋2)×12/8)＋511)/512

= 2 sectors

Once size 1 is calculated, at step S6, the file system operation unit 9 calculates NOM using Equation 12 so that the master boot record and partition boot area are located in different clusters.

NOM=(SC-(RSC+RDE+size 1×2)%SC)+SC

＝(2－(1+32+2×2)%2)+2

= 3 sectors

Once the NOM is calculated, the file system operation unit 9 calculates the number of clusters in the normal area using Equation 13 at step S7.

CN=(TS-(RSC+RDE+Size 1×2+NOM))/SC

＝(1280－(1+32+2×2+3))/32

＝620 sectors

In step S8, the file system operation unit 9 calculates the data length of the FAT using formula 14.

For FAT12: Size 2 = (((CN+2)×12/8)+511)/512

=(((620＋2)×12/8)＋511)/512

=2

In step S9, the file system operation unit 9 compares the size 1 calculated in step S5 with the size 2 calculated in step S8. Here, since the calculated size 1 and size 2 are both 2 sectors, it is judged that these two values are equal, and the process of this flowchart ends.

When the system area and partition control management area are composed of TS, size 1 and NOM calculated by the above calculation formula, the arrangement of the protection area is as shown in FIG. 18 . In this embodiment, both the user data area and the protected area include a partition control area, a system area and a protected area, so the access device 200 regards each area as an independent physical medium. In this way, although the sizes of the clusters contained in the user data area and the protection area are different, the boundaries of the user data area and the protection area can be consistent with the boundaries of the erasable block.

In the above embodiment, even though the volume area includes two areas—the user data area and the protection area, the boundaries of these areas coincide with the boundaries of the erasable blocks, so the cluster rewriting can be completed in a short time. Moreover, the number of erasable blocks that need to be erased is reduced, so the lifetime of the non-volatile memory is not shortened unnecessarily. third embodiment

A third embodiment relates to an improved method used in erasing data stored in an erasable block managed with logical addresses and physical addresses. The structure of the access control unit 2 in this embodiment is shown in FIG. 19 . In the figure, the access control unit 2 includes: an instruction decoding unit 21 , a logical/physical address conversion unit 22 , a read control unit 23 , an allocation conversion unit 24 , a write control unit 25 , an erasure control unit 26 and a modification control unit 27 .

The instruction decoding unit 21 receives the instruction issued by the access device 200, and decodes the original content of the instruction at the point of issuance. If a read instruction is received, the instruction decoding unit 21 instructs the read control unit 23 to read data from the nonvolatile memory 1 (read). If a write instruction is received, the instruction decoding unit 21 instructs the write control unit 25 to write data into the non-volatile memory 1 (write). If a write instruction designating a non-erased block as an access target is received, the instruction decoding unit 21 instructs the modification control unit to modify (rewrite) the data stored in the nonvolatile memory 1 (modify). If an acquire state instruction is received, the instruction decoding unit 21 instructs the allocation transformation unit 24 to read the list of erased blocks (the specification will be explained later). If an erasing instruction is received, the instruction decoding unit 21 instructs the allocation transformation unit 24 to erase the specified erasable block.

The logical/physical address conversion unit 22 includes an address correspondence table representing the correspondence between the logical and physical addresses of the erasable block. When the logical address to be accessed is determined by the access device 200, the logical/physical address conversion unit 22 refers to the corresponding relationship of the logical address shown in the address correspondence table, and converts the logical address into a corresponding physical address, and output the physical address to the read control unit 23, allocation transformation unit 24 and modification control unit 27.

When the access device 200 issues a read command, it is controlled by the write control unit 23 so that the access device 200 can read the data stored in the read position defined by the read command.

The allocation conversion unit 24 stores an erased block list, in which the physical addresses of erased erasable blocks are arranged in a first-in-first-out (FIFO) format. When the access device 200 issues a read command, the allocation transformation unit 24 judges whether the physical address corresponding to the logical address defined as the read target exists in the erased block list. If it is judged that the physical address exists, the allocation transformation unit 14 outputs the physical address to the write control unit 25, and deletes the physical address from the erased block list. If the erasable block indicated by the corresponding physical address has not yet been erased, then this allocation conversion unit 24 will distribute the physical address of the beginning of the erased block list to the logical address as the read target, and send to the read control Unit 25 outputs this allocated physical address, adding at the end of the erased block list the physical address previously allocated to this logical address as a read target. In addition, when the access device 200 sends an instruction to acquire status, the allocation conversion unit 24 converts the physical address shown in the erased block list into a logical address, and outputs a logical address table instruction to the access device 200. Erased block list of erased blocks. If an erase command indicating a logical address has been issued by the access device 200, the allocation transformation unit 24 will control the operation so as to add the physical address corresponding to the logical address into the erased block list.

When receiving a write instruction to perform a write operation on a block, the write control unit 25 writes data into the non-volatile memory 1 according to the physical address output by the allocation transformation unit 24 .

Erase control unit 26 polls at regular intervals to determine whether the physical address of an unerased block has been added to the erased block list. If the physical address has been added, the erase control unit 24 erases the erasable block indicated by the physical address.

When receiving a write instruction to modify an erasable block that has been written with data, the modification control unit 27 will read the data from the non-volatile memory 1 into the working memory 3, and keep the data in the working memory 3 to modify the data, and then write the modified data from the working memory 3 to the non-volatile memory 1. Such modification processing performed by the modification control unit 27 is accomplished through cooperative processing with the allocation conversion unit 24 . The flow of such data modification processing executed by the modification control unit 27 is shown in FIG. 20 in chronological order.

In the initial state shown in FIG. 20A , data is currently stored in the shaded blocks (the physical addresses of the blocks are 001, 002, 004, 007~015), and reserved blocks 003, 005, and 006 have been erased. Physical addresses 003, 005 and 006 representing erased blocks are arranged in the erased block list. In the address correspondence table representing the correspondence between logical addresses and physical addresses, logical addresses 001, 002, 003, 004, and 005 correspond to physical addresses 001, 002, 003, 004, and 005, respectively.

Here, if the write instruction issued by the access device 200 indicates that different values are written into the erasable block whose logical address is 001, then the write control unit 25 will write the data written in the block of logical address 001 from The non-volatile memory 1 is transferred to the working memory 3 (see arrow '①read' in FIG. 20A). Then the write control unit 25 modifies the data written in the blocks whose logical address is 001, and these blocks exist in the working memory 3 (see arrow '② modify' in FIG. 20A ).

Next in Fig. 2B, the modification control unit 27 reads out the previous physical address 003 in the wiped block list, as shown by the arrow BN1, and writes the data into the block indicated by the physical address 003 , as indicated by the arrow '③write'. Next in FIG. 20C , the modification control unit 27 arranges the physical address 001 of its corresponding block to be erased in the erased block list, as shown by arrow ④.

Finally, the modification control unit 27 exchanges the corresponding logical address/physical address, as shown in FIG. 20D . Here, physical address 003 is assigned to logical address 001, and physical address 001 is assigned to logical address 003. Therefore, by writing the data into the physical address 003, it is possible to rewrite the data into the logical address 001, and then perform corresponding conversion between the logical and physical addresses.

The internal structure of the access device 200 will be explained below. FIG. 21 shows the internal structure of the file system operating unit 9 in the access device 200 of the third embodiment. The file system operation unit 9 shown in the figure includes: a file deletion unit 28 and a priority erasure control unit 29 .

The file deletion unit 28 deletes files by updating the FAT read from the main memory 7 , and writes the updated FAT into the semiconductor memory card 100 . If the application program 8 sends an instruction to delete a file, then the file deletion unit 28 will delete the file by setting the FAT entry corresponding to the corresponding cluster that stores the file fragment to '0', '0' means not The clusters that are used. However, these clusters (erasable blocks) storing file fragments have not yet returned to a blank state, so these clusters must be erased at first before other files can be stored in these clusters. Therefore, the time required for erasing processing increases in proportion to the increase in the number of times old files are deleted and new files are recorded.

While waiting for an instruction from the application program 8, the priority erasure control unit 29 issues an acquisition status instruction to the semiconductor memory card 100, ordering it to read the erased free memory list. Once the command is issued and the erased free memory list is output, the priority control unit 29 receives the outputted erased free list, and compares the list with the FAT, so as to determine the Erasable blocks corresponding to multiple clusters that are set to be unused. (When an erasable block contains multiple clusters, only those erasable blocks composed of unused clusters can be determined). Therefore, among these erasable blocks, those blocks that do not exist in the list of erased blocks are also determined. These blocks are erasable blocks that are neither used nor erased, and the priority erasure control unit 29 issues an erase instruction to the semiconductor memory card 100 to command erasure of these blocks. The erasing command includes a logical address definition, and a command to add the physical address corresponding to the logical address into the list of erased blocks. When polling is performed by the erasing control unit 26, if a physical address is added to the erased block list by issuing such an erasing command, then the erasing method of the block indicated by the erasing command is the same as Other unused and unerased blocks are treated in the same way. And the same process is performed for all unused and unerased blocks. The detailed process performed by the priority erasing control unit 29 will be described below with reference to the flowchart of FIG. 22 .

In this embodiment, if the priority erasing control unit 29 is activated, it will go to the loop processing of steps S21-S22. In step S21, the priority erasing control unit 29 judges whether the access command issued by the application program 8 has been received, and in step S22, judges whether the predetermined polling time has elapsed. If an instruction from the application program 8 has been received, the priority erasure control unit 29 goes to step S23 to perform file system operations and access the semiconductor memory card 100 according to the instruction. If the predetermined polling time has passed, then this priority erasing control unit 29 turns to step S24 from step S22, and sends an acquisition status command to the semiconductor memory card 100 to read out the free list that has been wiped, and then, at Step S25, the priority erasing control unit 29 is in a waiting state, waiting to read out the list of erased blocks. Once the list of erased blocks is read out, in step S26, the priority erasure control unit 29 is used for all erasable blocks made up of unused clusters in the FAT and the erasable blocks that do not exist in the list of erased blocks. A wipe block defines a logical address.

In step S27, the priority erasure control unit 29 judges whether there are unused erasable blocks that have not yet been erased, and if there is no such block, immediately return to the loop processing of steps S21-S22. If there is such a block, the priority erasure control unit 29 executes the loop processing of steps S28-S30. In other words, the priority erasure control unit 29 issues an erasure command to the semiconductor memory card 100 to ensure erasure of data at each physical address of the unused and unerased block determined in step S26. Once such an instruction is issued, the physical address of each such block is added to the erased block list and erased.

The operation of the above-mentioned priority erasure control unit 29 will be described in more detail below with reference to FIGS. 23A and 23B. In the initialization state shown in FIG. 23A, an erasable block address 0001 is set to be in use in the FAT, and physical addresses 0003, 0005, and 0006 are arranged in the erased block list as erased block addresses . Data is stored in blocks of the file system shaded in the figure (addresses 0000, 0001, 0002, 0004). In this state, if there is no instruction from the application program 8, and the predetermined polling time has passed (no in step S21, yes in step S22), the file system operating unit 9 in the access device 200 sends The semiconductor memory card 100 issues an acquire status command to read the erased block list (step S24), and compares the read erased block list with the FAT (step S25). The result of the comparison is to determine those blocks that are set as unused in the FAT, and those blocks whose logical addresses do not exist in the erased block list (the erasable block address is 0000, 0002, 0004) (step S26). Since these blocks are not in use, the file operating system unit 9 sends an erase command to the semiconductor memory card 100 to erase these erasable blocks (steps S28-S30). Therefore, as shown in FIG. 23B, addresses 0000, 0002, and 0004 are added to the erased block list, and the erasable blocks indicated by these addresses are erased.

In the above embodiment, the access device 200 reads the erased block list from the semiconductor memory card 100, and compares the read erased block list with the FAT, so as to determine the unerased blocks that have not yet been erased. used erasable blocks, and command the semiconductor memory card 100 to erase these blocks. So, while waiting for an instruction from said application program 9, the file operating system unit 9 can erase those unused erasable blocks which are still not erased and make this erasure of these unused blocks more efficient to execute.

In addition, erasure of unused blocks can be performed whenever the access device 200 is idle, so the chances of performing block erasure can be greatly increased. Therefore, there is no danger of insufficient physical addresses in the erasable block list, and the processing efficiency when modifying and rewriting blocks will be greatly improved.

In the access device 200 in this embodiment, the erased block list is read out from the semiconductor memory card 100, and unused and unerased blocks are determined. However, in the access control unit 2 of the semiconductor memory card 100, the erasure control unit 26 may determine blocks that are neither used nor erased according to the FAT, and then delete these blocks. Here, the access control unit 2 converts the physical address in the erased block list into a logical address, and transmits the converted address to the access device 200 . However, it is also possible to pass the erased block list and the address correspondence table to the access device 200, in which the physical address is converted into a logical address. Data erasure can be performed while deleting a file, or as a parallel process while executing other commands. Also, other information such as an unerased block list managing physical addresses of non-erased erasable blocks may be used instead of an erased block list managing physical addresses of erased erasable blocks. This embodiment is described using FAT, but the information can be stored in a list or other similar forms. Fourth Embodiment

The fourth embodiment illustrates an improvement achieved by performing the processing of the third embodiment with an erased block list instead of an erased block list. Fig. 24 shows the internal structure of the access device 2002 of the fourth embodiment. If this figure is compared with Figure 19, it can be seen that the erased block list has been replaced by an erased block table. The erased block table in the fourth embodiment consists of a plurality of entries corresponding to each erasable block. If a block is erased, the corresponding entry is set to '1', and if a block is not erased, the corresponding entry is set to '0'. The erased block table represents the erasing state of each erasable block by setting '1' or setting '0' in the erased block table.

Since the erased block list has been replaced by the erased block table, the processing performed by the write control unit 25 and the erase control unit 26 in the fourth embodiment is the same as that in the third embodiment in the following respects. Execution is different.

When the access device 200 generates a write command, the write control unit 25 in the fourth scheme judges whether to erase the write target determined by the command according to the erased block table, and judges whether it is an unused of blocks. In this way, if the write target block is an unerased block, before writing new data into the block, the write control unit 25 instructs the erasure control block 26 to erase the data in the block. In other words, the write control unit 25 in the fourth embodiment will erase data before writing data.

Next, the processing procedure of the priority erasure control unit 29 in the fourth embodiment will be explained. While waiting for an instruction from the application program 8, the priority erasure control unit 29 performs erasure of erasable blocks. This erasing process is the same as described in the third embodiment, and is shown in the flowchart of FIG. 25 . The flow chart is consistent with that shown in Figure 22 except that the "erased block list" has been replaced with an "erased block table". And since the content has not changed much, a detailed explanation is omitted here. The processing performed by the priority erasing control unit 29 in the fourth embodiment will be explained below with reference to FIGS. 26A and 26B.

In the initial state shown in Figure 26A, the block whose address is 0001 is managed as a used block, and the blocks whose addresses are 0003, 0005, and 0006 in the erased block table are managed as blocks of erased data . Data is stored in blocks of non-volatile memory that have been shaded in this figure (addresses 0000, 0001, 0002, 0004). In this initial state, if the blocks are unused but contain data, and a write command to write data to these unused blocks has been generated, data erasing processing needs to be performed. In this embodiment, if the priority erasing control unit 29 is waiting for the application program 8 to issue an instruction, and the predetermined polling time has passed (no in step S21, yes in step S22), then the priority erasing control unit 29 The erased block table is read by sending an acquire state command to the semiconductor memory card 100 (step S24), and the read erased block table is compared with the FAT (step S25). The result of this comparison is to find out blocks that are set as unused in the FAT and whose logical addresses are not managed by the erased block table (the erasable block addresses are 0000, 0002, 0004) (step S26). In FIG. 26, for the blocks with addresses 0000, 0002 and 0004, '0' is set in the FAT and the erased block table, so these blocks are not used and remain as unerased blocks. Since these blocks are unused, the contained data can be safely erased, so the priority erase control unit 29 issues an erase command so that each corresponding address in the erased block table is set to '1' , to indicate that these blocks will be erased (steps S28-S30). Therefore, all data in unused blocks are erased, and then the erased block table manages these blocks as erased blocks. Therefore, data writing processing can be performed without any erasing processing by allocating unused blocks, which will be described later in this specification, and high-speed data writing can be realized. fifth embodiment

The fifth embodiment proposes a method for solving the problem of segmented file storage. Conventionally, file fragmentation refers to the process of dividing a file into a plurality of file fragments, and then storing these fragments in a memory discontinuously. If a plurality of file fragments are stored discontinuously on a recording medium, such as an optical disc or a magnetic disk, it is necessary to search a large number of disks to read out the discontinuously stored file fragments, thus prolonging the reading processing time. In order to solve the problem caused by such segmented storage, the file system operation unit 9 reads out the non-contiguously stored file segments, and then stores them continuously in the memory (eliminates segmented storage). If the above-mentioned processing is completed, file fragments constituting one file can be smoothly read out without performing any disk seek, and data can be read out at high speed. This summarizes the solution to the multi-disk fragmentation problem. Fragmented storage that occurs in the semiconductor memory card 100 will be considered below. When data is read out from the semiconductor memory card 100, no disk seek is required, so if the file segments are stored non-sequentially, it is impossible to lengthen the processing for this reason.

However, although disk seeks are not a problem, segmented storage will still result in a substantial increase in 1. additional operations when sending commands; 2. cache misses when reading FAT (both cases will be explained below detailed description); thus the time required to read a file is increased.

1. Additional operations that occur when sending commands

It is easier to understand the impact of segmented storage on additional operations when instructions are sent if it is described with reference to FIGS. 27A to 27B . FIG. 27A exemplarily illustrates a situation where a 48KB file is divided into three 16KB file fragments and randomly stored in three areas. Fig. 27B exemplarily illustrates a situation: a 48KB file is stored in a continuous storage area. In FIGS. 27A and 27B, the file system operation unit 9 issues a read command in the format of 'read (address, size)' to the nonvolatile memory to execute reading data. Here, 'address' represents the logical address of the read target, and 'size' represents the size of the data to be read. The time spent for each read command is equal to the sum of the read data time, which is proportional to the length of the read data and a certain additional time required to send the read command. In the data arrangement shown in FIG. 27A, reading file 1 must send three read commands: 'read (0000, 16KB)', 'read (0002, 16KB)' and 'read (0004, 16KB)'. This means that the time required to read out the file is 36 ms (=(8+4)×3). On the other hand, if the data is arranged according to Fig. 27B, then reading all 48KB data of the file only needs to send a read command simply: 'read (0000, 48KB)'. Therefore, in this example, the time required to read the file is 28ms (=(8*3+4)). In this way, it can be seen that the additional operations generated when sending commands are proportional to the randomness of file fragment storage.

2. FAT cache

FAT cache is a look-ahead read operation in which the FAT arranged in the system area of the semiconductor memory card 100 is read into the internal memory of the access device 200, and the access device 200 allows high-speed query by accessing its internal memory The FAT. If the access device 200 is a portable device with a small-scale internal memory, the FAT needs to be cached frequently, and the FAT is cached by sequentially reading FAT segments continuously stored in the memory. However, if the FAT has been segmented, caching the FAT will read both the segment containing the FAT and the segment containing other data into the internal memory in sequence. If the data is only read into the internal memory without reading the FAT segment, a large amount of data that is not a cache target will be read, and a cache miss will occur. If a large number of such cache misses occur, the number of reads to the internal memory will increase accordingly, thereby prolonging the time required to read the FAT.

Despite these two problems, it is unwise to solve the fragmentation problem in the semiconductor memory card 100 in the same way that it is solved in the magnetic disk. The reason for this is that to modify the data stored in the semiconductor memory card block, the stored data needs to be erased, and the erasing takes longer than the time required to perform the same operation on a magnetic disk. processing time. In this embodiment, it is recommended to change the corresponding relationship between the physical address and the logical address as a countermeasure to prevent the additional operation time of the instruction and the occurrence of a cache miss during segmented storage. In other words, by changing the correspondence between physical addresses and logical addresses, blocks storing FAT segments can be continuously represented by logical addresses without changing their physical arrangement in the memory.

As shown in Figure 27A, when a 48KB file is divided into three 16KB segments and stored in three independent storage areas, when these blocks are continuously indicated by logical addresses, the access device 200 can read File fragments stored in these three areas (blocks). The read operation is performed by sending a read command to indicate the first address of these logical addresses and the length of the file (here is 48KB). In addition, if the FAT is divided into 3 segments and stored in three storage areas (blocks), then, as long as the logical addresses of these blocks are continuously indicated, the access device 200 can store these discontinuously stored FATs Fragments are sequentially read into the internal memory. Here, the read operation is performed by issuing a read instruction sequentially designating each consecutive logical address as a read target. If this processing is performed, cache misses no longer occur. Therefore, even if multiple files and FAT fragments are randomly stored in multiple areas, since continuous logical addresses can be assigned to these areas, the problem of additional operations generated when issuing read instructions is solved, and the problem of cache misses can also be avoided. produce.

The operation of the priority erasing control unit 29 applying logical addresses to solve the segmented storage problem will be described with reference to the flow chart shown in FIG. 28 . The explanation will also refer to Figure 29, which shows the actual objects to which the variables t, u, v and y correspond in the flowchart. In step S50, the priority erasing control unit 29 assigns initial values (x←1, y←1) to variables x and y. This represents the processing done at the first segment of the first file. Next, in step S51, the priority erasure control unit 29 compares the logical address s corresponding to the block storing the yth fragment of the xth file with the logical address t corresponding to the block storing the y+1th fragment of the xth file Compare to determine if these blocks are adjacent. As shown in Figure 29A, when the y-th segment and the y+1-th segment are stored in non-adjacent erasable blocks, and the order is y→y+1, step S52 is "Yes", and the priority erasing control unit 29 Go to step S54.

As shown in FIG. 29B, when the storage order of these segments is y+1→y, the priority erasure control unit 29 turns to step S53, and assigns physical address u to logical address t, and assigns physical address v to logical address s. Accordingly, this change in assigning physical addresses to logical addresses is shown by arrows rv1 and rv2 in FIG. 29B. Next, in step S54, the priority erasure control unit 29 judges whether the logical address of the erasable block storing the y-th segment is continuous with the logical address of the erasable block storing the y+1-th segment. As shown in Figure 29C, if the erasable block storing the y+1th segment is followed by the erasable block storing the yth segment, then the relational expression s+1=t is established, so step S54 is "yes", and the processing is transferred to Go to step S56. If the erasable block storing the y-th segment is not adjacent to the y+1-th segment, then the relational expression s+1=t is not established, step S54 is "No", and the processing goes to step S55. Here, if the physical address corresponding to the logical address s+1 is w, then in step S55, the physical address v of the erasable block storing the y+1th segment is assigned to the logical address s+1, and the physical address w is assigned to the logical address t. As shown in Fig. 29D, when the y-th and y+1-th fragments are stored, if the process of step S55 is executed, the corresponding relationship of addresses will change as indicated by arrows rv3 and rv4.

Once the above processing is completed, the priority erasure control unit 29 makes a judgment in step S56: whether the variable y is the number of the final segment. If y is not the final number, the priority erasure control unit 29 increments y by 1 at step S57, and proceeds to step S51. By performing the check in step S56, repeat the processing of S51 to S57 for each segment of the xth file, and add 1 to y in step S57.

In other words, if step S56 is "yes", then the priority erasure control unit 29 judges whether x is the last number of the file in step 58, and if the answer is "no", then go to step S59, add variable X to 1(x←x+1), and assign an initial value to variable y in step S60. Therefore, by performing the check of step S58, the processing of steps S51 to S57 is repeatedly performed for all files, and 1 is added to the variable y in step S59.

The solution to the problem caused by segmented storage proposed in this embodiment will be described with reference to the example shown in FIG. 30 . In the initial state shown in FIG. 30A, the fragment files 1-1/3, 1-2/3 and 1-3/3 constituting file 1 are stored in erasable blocks corresponding to logical addresses 0000, 0001 and 0004. In addition, the fragment files 2-1/4, 2-2/4, 2-3/4, and 2-4/4 constituting file 2 are stored in erasable blocks corresponding to logical addresses 0002, 0003, 0005, and 0006.

Store the corresponding address (logical address 0002) of the erasable block of the segment file 2-1/4 and then store the logical address 0001 of the segment file 1-2/3, and store the logic corresponding to the erasable block of the segment file 1-3/3 The address is 0004.

If the processing method of Fig. 28 flow chart is applied to the embodiment of this figure, then segment file 1-1/3 and segment file 1-2/3 correspond to logic address 0000 and 0001 respectively, like this, the sequence of arrangement of these segments just The order is the same as that of the corresponding logical address, and the corresponding relationship does not need to be exchanged. However, if you notice the relationship between fragment files 1-2/3 and 1-3/3, you can see that when the erasable block storing fragment file 1-2/3 corresponds to logical address 0001, it corresponds to The next erasable block at logical address 0002 will be file 2-1/4. Therefore, "NO" in step S54, and the process goes to step S55. Then, in step S55, the corresponding relationship between the logical address and the physical address of the erasable block of the storage segment file 1-1/3 will be compared with the logical address and the physical address of the erasable block of the storage segment file 2-1/4 The corresponding relationship between them is exchanged. Therefore, fragment files 1-2/3 and 1-3/3 can be represented by continuous logical addresses, so logical address 0004 is assigned to physical address 0002 of the erasable block storing fragment file 2-1/4, and vice versa , the logical address 0002 is allocated to the physical address 0004 of the erasable block storing the segment file 1-3/3. After this change, the corresponding relationship between the physical address and the logical address is shown in FIG. 30B. Once the processing of the segment file 1-3/3 is completed, the processing of the segment file 2-2/4 is performed. In Fig. 30B, the relationship between segment files 2-1/4 and 2-2/4 is: segment file 2-1/4 is stored in the erasable block corresponding to logical address 0004, and the next segment file 2- 2/4 is stored in the erasable block corresponding to logical address 0003. Thus, the logical addresses attached to these fragments should be in reverse order. Therefore, "NO" in step S52, and the process goes to step S53, where the order of addresses corresponding to segment file 2-1/4 and file 2-2/4 is interchanged. Here, the physical address 0002 corresponding to the erasable block storing the segment file 2-1/4 is assigned to the logical address 0003, and the physical address 0003 corresponding to the erasable block storing the segment file 2-2/4 is assigned to the logical address Address 0004. The result of such processing is that the file fragments constituting file 1 and file 2 are all indicated by continuous logical addresses.

In this embodiment, even if the file fragments are stored in discontinuous storage areas, continuous logical addresses can be assigned to these storage areas, which reduces the additional operations generated when issuing read commands and the cache FAT and File cache misses. Sixth Embodiment

The sixth embodiment sets forth an improved method applied to recording different contents in said non-volatile memory with a distribution service, and reproducing the recorded contents on a portable playback device. In this case, files having various attributes such as music, images, games, and texts are to be stored in the semiconductor memory card 100 . If there is a difference in whether the file area is readable or editable according to the file type, then the file system operating unit 9 in the access device 200 needs to check the contents of the file before reading or editing. If such a check is required every time a file is read or edited, the processing performed by the file system operation unit 9 becomes complicated. Therefore, the universal disc format (UDF) is used in the digital versatile disc (DVD) to set an extended attribute for each file, so that the type of data stored in the file can be recognized instantly without viewing the content of the file.

Figure 31 shows an embodiment of an extended attribute. The attribute shown in the figure is a new type of extended attribute formed according to the implementation purpose extended attribute in the UDF. The extended attribute includes: attribute type 2000, attribute subtype 2010, reserved area 2020, attribute length 2030, implementation usage length 2040, implementation identifier 2050 (the same as the field of implementation usage extension attribute in the UDF, so omitted here more detailed description) and implementation use 2060. The implementation usage 2060 includes: a header checksum 2061 to store a checksum for the header part of the extended attribute, a name 2063 to store a file name, and a flag 2062 to store a file attribute. Set each bit of the flag 2062 to '0' or '1' to represent the file attribute. The first bit of the flag 2062 indicates whether the corresponding file is a text file, the second bit indicates whether the corresponding file is an image file, and the area above the third bit is a reserved area.

If the extended attributes are stored in the semiconductor memory card 100, a difficulty is apparent. That is, as mentioned above, the volume management information of the semiconductor memory card 100 complies with the data format proposed by the ISO/IEC9293 standard, so it is not stipulated to give each file an extended attribute.

In this embodiment, the extended attributes are stored in at least one of the protected area and the backup area. Since the protection zone has already been explained in the second embodiment, further explanation is omitted here. The backup area is managed independently of the normal area in the user area. When a bad sector appears in the user area of the nonvolatile memory, the allocation transformation unit 24 selects a sector from the backup area to replace the bad sector. FIG. 32 shows the internal structure of the semiconductor memory card 100 including bad sectors. Such attributes can be set for each file by storing new extended attributes that cannot be defined in the FAT file system in the protected area or the backup area. Furthermore, since the user area and the user data area used by the usual users use an ordinary file system completely irrelevant to the newly set extended attributes, the system is compatible with other systems. In addition, since the access device 200 only needs to perform operations necessary for each type of file, the access device 200 performs fewer operations, thereby reducing the memory size of the access device 200 . Furthermore, since the file type of each file can be judged only by means of the extended attribute information and not necessarily by means of the content of the file, high-speed operation can be realized. Seventh embodiment

This implementation scheme proposes setting the files in the FAT file system as prohibition of writing and prohibition of reading, so as to further strengthen the protection of the files. In a common FAT file system, the attribute of an independent file can be set as write-prohibited and read-prohibited in the file entry. When the access device 200 is connected to the semiconductor memory card 100, the file system operation unit 9 reads out and saves the volume management information, and judges whether the file can be read by means of the file attribute of a specific file. or write. If the application program 8 in the access device 200 accesses the semiconductor memory card 100 through the file system operation unit 9, the attribute in the file entry is valid. However, if the application program 8's access to the semiconductor memory card 100 is to skip the file system operation unit 9, and to perform writing or reading by sending a direct write or read instruction to the semiconductor memory card 100, then in the Any write-disable or read-disable attributes set in the above file entry will have no meaning. Here, this embodiment proposes to construct the semiconductor memory card 100 as shown in FIG. 33 so that even if the application program 8 directly accesses the semiconductor memory card 100, write prohibition and read prohibition will be effective. FIG. 33 shows the internal structure of the semiconductor memory card 100 in the seventh embodiment. The feature of this figure is that the non-volatile memory 1 has a backup area including a block attribute table.

The block attribute table is composed of entries corresponding to erasable blocks in the volume area. If an entry is set to '1', the corresponding block is write-prohibited. Also, if an entry is set to '2', the corresponding block is read disabled. The file system operation unit 9 and the access control unit 2 explained in the fifth embodiment perform operations according to block attribute tables in addition to operations according to file entries. If the application program 8 instructs the file system operation unit 9 to open the file with the attribute set, the file system operation unit 9 will set the attribute in the file entry corresponding to the file. For example, if the application program 8 has received an instruction to set a certain file as write-prohibited, then the attribute of the corresponding file entry is set as write-prohibited. If the application program 8 has received an instruction to set the file as read-prohibited, then the attribute of the corresponding file entry is set as read-prohibited. If the application program 8 has received an instruction to set the hidden attribute, the file system operation unit 9 sets the read prohibit attribute in the corresponding file entry.

If the file attribute is set in the file entry in this way, the file system operation unit 9 divides the file into cluster-sized fragments, and records these fragments in a plurality of clusters in the user area. After recording these file fragments in the cluster, the file system operation unit 9 sets the file attribute in the entry corresponding to each erasable block storing the file in the block attribute table. If the file is prohibited from being written, then the file system operating unit 9 will set the write-prohibited attribute in the entry corresponding to each cluster storing the file segment; if the file attribute is read-prohibited, then in the The read prohibition attribute is set in the entry corresponding to each cluster storing the file fragment. If the file attribute is hidden, the file system operation unit 9 sets the read prohibition attribute in the entry corresponding to each cluster storing the file fragment.

If the file is recorded in the user area in this way, data is read or written to the block according to the attributes shown in the block attribute table. In other words, if the access device 200 issues a read command, the read control unit 23 will refer to the entry in the block attribute table corresponding to the target address. If the entry indicates that it is allowed to read, then the read control unit 23 will read the data from the block indicated by the read target address, otherwise, if the entry indicates that reading is prohibited, then it will not be read from the block indicated by the read target address. Read out the data.

If the access device 200 issues a write instruction, the write control unit 25 or the modification control unit 27 consults the entry corresponding to the write target address in the block attribute table. If the entry indicates that writing is allowed, the write control unit 25 or the modification control unit 27 writes data into the block indicated by the write target address, and if the entry indicates that writing is prohibited, the data is not written into the block. Write to the block indicated by the target address.

This writing and reading control can be executed through the file system operation unit 9 when the application program 8 issues a write or read command, or can be directly issued by the application program 8 to write or read the command by skipping the file system operation unit 9 Executed when the command is read. Therefore, no matter whether the command issued by the access device 200 passes through the file system operation unit 9, the reading and writing of blocks can be restricted. The file system operation unit 9 in this embodiment will be explained with reference to the example in FIG. 34 . In Figure 34A, the data contained in file 1 is stored in erasable blocks with addresses 0000, 0002 and 0003, the data contained in file 2 is stored in erasable blocks with addresses 0001 and 0004, and the data contained in file 3 is stored in In the erasable block whose address is 0005, the data contained in the file 4 is stored in the erasable block whose address is 0006. Fig. 34B shows the state when file 1 is stored, which has the write inhibit attribute set. In the block attribute table, the entries corresponding to the blocks (addresses 0000, 0002 and 0003) storing the fragments of file 1 are set to '1', indicating the write prohibition attribute. If the block attribute table is set in this way, the write control unit 25 will refuse to execute the write operation command on the blocks whose addresses are 0000, 0002 and 0003.

FIG. 34C shows the state when the file 2 is stored in the non-volatile memory 1, and the read prohibition attribute has been set for the file. In this example, the flag indicating the read prohibition attribute is set in the entry corresponding to file 2 in the block attribute table. Then, the file 2 is divided into a plurality of fragments, and these fragments are stored in a plurality of blocks in the user area.

When the file segments of file 2 are stored in blocks with logical addresses 0001 and 0004, the entries corresponding to these blocks in the block attribute table are set to '2', indicating the read prohibition attribute. If the block attribute table is set in this way, the read control unit 23 will refuse to execute the instruction to read data from the blocks whose addresses are 0001 and 0004.

In the above-described embodiment, the read-prohibited and write-prohibited attributes can be set corresponding to each block in the semiconductor memory card 100, so even if the application program 8 skips the file system operation unit 9 and directly accesses the semiconductor memory Card 100, the semiconductor memory card 100 can also prevent access to prohibited files. Therefore, by setting the read-prohibited and write-prohibited attributes on the block storing the file in this way, the protection of the file can be guaranteed, and if the file is copyrighted, copyright protection can also be realized.

In this embodiment, the flag representing the write-prohibited attribute is represented by "1", and the flag representing the read-prohibited property is represented by "2", but these are only examples, and the present invention is not necessarily limited by such flags. limit. In addition, in this embodiment, the block attribute table is used to attach read and write prohibition attributes to each block in the non-volatile memory 1, but if an attribute is to be set for each individual block, it can also be used list or similar structure. In the present embodiment, the method explained is for setting the read and write prohibition attribute as the block attribute in the nonvolatile memory 1, but information other than those described in these embodiments may also be used as the block attribute. Embodiments include block management, so only users with fixed privileges can gain access; or give each block a user ID, and only allow users with that ID to have access; or set block access rights individually for each user . In the above description of the file system, a FAT file system was used, but similar effects can be obtained using other conventional file systems such as UDF or the New Technology File System (NTFS) used in WindowsNT ^™ or a custom file system. In addition, in these embodiments, the number of sectors included in an erasable block is 32, but this is just an example, and the number of sectors in an erasable block may be more or less than this number.

Industrial applicability

As described above, the semiconductor memory card 100 in the present invention prolongs the life of the nonvolatile memory 1 . Therefore, when using this semiconductor memory card 100 to record music contents obtained through digital music distribution, even if the recording is performed repeatedly, that is, these music contents are erased first, and then other music contents are frequently recorded, there will be no problem. The lifetime of the nonvolatile memory 1 is shortened. Therefore, a single semiconductor memory card 100 can be used to repeatedly record music contents obtained through digital music distribution.

Claims (25)

1. access means to semiconductor memory card execute file visit, described semiconductor memory card has a memory block of being made up of a plurality of sectors, and described access means is then passed through every group 2 ^jData in (j be 0 or positive integer) individual sector manage as one bunch, and one or more bunches managed as a file carry out described visit, in the described memory block every group 2 ⁱIndividual contiguous sector is formed a piece (i be 0 or positive integer), and piece is the least unit that can carry out data erase, and this access means comprises:

Computing unit, can be operated be used for according to manage in the described memory block bunch quantity calculate the size of volume management information, described volume management information comprises a file allocation table, is used to each file to indicate link between each bunch of corresponding this document;

Reserve the unit, can be operated and be used for reserving (1) first area and (2) second area that are used for user data that are used to write down described volume management information, the data length that described first area had is greater than the volume management information of being calculated, and by m * 2 ^j(m is a positive integer) individual sector is formed, and second area is made up of the sector of following the first area; And

Record cell, can be operated the record volume management information that is used in described first area, in second area user data and with these volume management information and user data as bunch managing.

2. the access means of claim 1, wherein:

Except described file allocation table, described volume management information also comprises Main Boot Record, partition table, subregion boot section information and root directory entry; And

Described record cell can be operated and be used for writing down described Main Boot Record and partition table in first sector of described first area, and after skipping the sector of predetermined quantity, in follow-up sector, write down described subregion boot section information, file allocation table and root directory entry, so that make the end of described first area consistent with the end of root directory entry.

3. the access means of claim 2, wherein:

Described computing unit can be operated and be used for calculating itself and SUM by the sector number addition that will be used to write down described subregion boot section information, file allocation table and root directory entry;

Described reservation unit can be operated be used for utilizing according to the m value that formula 1} calculates reserves the first area,

{ formula 1}NOM+SUM=2 ^j* m,

NOM is a sector number; And

Described record cell can be operated and be used for calculating described predetermined sector number by sector number NOM being subtracted 1.

4. the access means of claim 3, wherein:

Described record cell can be operated the sector that is used for being provided with described predetermined quantity in the described partition table of the described volume management information of record.

5. the access means of claim 3, wherein:

Described file allocation table have a plurality of entrys, with corresponding some bunches of some entrys, and every entry has been represented in identical file the link to other bunch;

Described access means comprises:

Receiving element can be operated the setting that is used for being received in the described memory block the sector sum;

Described computing unit comprises:

First computing unit, can be operated be used for by with described sector sum divided by sector number 2 ^jCalculate total number of clusters, and can be operated and be used for calculating the size of described file allocation table by the bit length that described total number of clusters be multiply by an entry, and

Described reservation unit comprises:

Second computing unit can be operated and is used for by the size of described file allocation table and the sector number addition that constitutes described partition table and root directory entry respectively being calculated itself and SUM, and should { formula 1} calculates the m value with the SUM substitution.

6. the access means of claim 5, wherein:

Described memory block comprises: have only the protected location that could conduct interviews by this equipment when the equipment that links to each other with described semiconductor memory card authorized when checking and

No matter whether the mandate of described associated devices is verified the equal user data area that can be conducted interviews by this equipment;

Described receiving element can be operated the setting that receives the source that is used for outside described access means the sector number of distributing to each described protected location and user data area;

Described first computing unit can be operated the size that is set at each described protected location and user data area calculation document allocation table that is used for according to being received; And

Described second computing unit can be operated and be used for respectively by each in the protected location is big or small and the user data area size and the sector number addition that constitutes the sector number of described subregion boot section information and constitute the root directory entry; so that calculate two and SUM of this protected location and user data area respectively; { formula 1} calculates the m value of this protected location and user data area respectively with each value substitution of these two and SUM again.

7. the access means of claim 2, wherein said record cell can be carried out:

(a) first handle: the user data that will be divided into a plurality of fragments is recorded as some bunches in described second area, and each fragment is stored in some bunches the cluster in this second area:

(b) second handle: a plurality of entrys are set, a link between each entry indication bunch in described file allocation table; And

(c) the 3rd handle: the positional information of filename of record and first bunch of position of expression this document in described root directory entry.

8. the access means of claim 1, wherein:

Described reservation unit can be operated and be used for reserving a backup area except that first and second zones in described memory block; And

When the user data in being recorded in described second area managed as the file with extend property, described record cell can be operated the attribute information that is used for writing down the expression extend property in this backup area.

9. the access means of claim 1, wherein:

Described semiconductor memory card also writes down the attribute list that comprises many attribute entrys, and every entry is corresponding to one of some;

Being recorded in user data in the described second area has as at least one and writes the file of forbidding attribute and manage; And

Described record cell can be operated to be used for writing for the following entry setting in the described attribute list forbids attribute, promptly described entry corresponding to fully by with write the piece that the corresponding sector of each bunch of forbidding in the file is constituted.

10. the access means of claim 1, wherein:

Described volume management information comprises a file allocation table;

Be set to first value corresponding to the entry that does not use bunch in the described file allocation table, and be set to the value of non-first value in the described file allocation table corresponding to the entry of usefulness bunch;

Described semiconductor memory card also preserves the piece of a wiping tabulation that contains physical address, and this physical address has been specified the piece of wiping in a plurality of of this memory blocks; And

This access means comprises:

Designating unit, it can be operated to be used to refer to fixs the piece of stating: (a) by bunch in all that in described file allocation table, be set to the piece do not formed with the sector; (b) be not changed to the piece of wiping described the wiping in the piece tabulation; And

Erase unit, it can be operated the piece that is used for wiping appointment.

11. the access means of claim 1, wherein:

Described volume information comprises a file allocation table;

Be set to first value corresponding to the entry that does not use bunch in the described file allocation table, and be set to the value of non-first value in the described file allocation table corresponding to the entry of usefulness bunch;

Described semiconductor memory card also preserves the piece of a wiping table that contains physical address, and this physical address has been specified the piece of wiping in a plurality of of this memory blocks; And

This access means comprises:

Designating unit, it can be operated to be used to refer to fixs the piece of stating: (a) by bunch in all that in described file allocation table, be set to the piece do not formed with the sector; (b) be not changed to the piece of wiping in the piece table described the wiping; And

Erase unit, it can be operated the piece that is used for wiping appointment.

12. the access means of claim 1, wherein:

Described user data is divided into a plurality of fragments, and each fragments store is in a piece;

Described semiconductor memory card comprises a corresponding tables, physical address under each piece of a fragment of expression storage and the corresponding relation between the logical address; And

When this corresponding tables indicates a plurality of physical address correspondences of a plurality of that are used for the storaging user data fragment when a plurality of discontinuous logical address,

This access means comprises:

Allocation units can be operated and are used for giving some continuous logical addresses with these physical address assignments;

Instruction sending unit can be operated and is used for reading by sending the user data fragment that reads instruction in these pieces to described semiconductor memory card, and this reads instruction and has specified the start address of these continuous logic addresses.

13. one kind with the stored program recording medium of computer-readable format, is used for initializing computer, so that by with every group 2 ^jData on (j be 0 or positive integer) individual sector manage as one bunch, and one or more bunches managed as a file and execute file visit on a semiconductor memory card, described semiconductor memory card has a memory block of being made up of a plurality of sectors, in this memory block, every group 2 ⁱIndividual contiguous sector is formed a piece (i be 0 or positive integer), and piece is the least unit that can carry out data erase, and this recording medium makes computing machine carry out following steps:

Calculation procedure, according to manage in the described memory block bunch quantity calculate the size of volume management information, described volume management information comprises a file allocation table, is used to linking between each file indication and corresponding each bunch of this document;

Reserve step, reserve (1) first area that is used to write down volume management information, and (2) second areas that are used for user data, the data length that described first area had is greater than the volume management information of being calculated, and by m * 2 ^j(m is a positive integer) individual sector is formed, and described second area is made up of the sector of following the first area; And

Recording step, record volume management information in the first area, user data in second area, and with these volume management information and user data as bunch managing.

14. the recording medium of claim 13, wherein:

Except described file allocation table, described volume management information also comprises Main Boot Record, partition table, subregion boot section information, root directory entry; And

Described recording step writes down described Main Boot Record and partition table in first sector of described first area, and after skipping the sector of predetermined quantity, in follow-up sector, write down described subregion boot section information, file allocation table and root directory entry, so that make the end of described first area consistent with the end of root directory entry.

15. the recording medium of claim 14, wherein:

Described calculation procedure calculates itself and SUM by the sector number addition that will be used to write down described subregion boot section information, file allocation table and root directory entry;

{ the m value that formula 1} calculates reserves the first area to described reservation step by basis

{ formula 1}NOM+SUM=2 ^j* m,

NOM is a sector number; And

Described recording step calculates described predetermined sector number by sector number NOM being subtracted 1.

16. the recording medium of claim 15, wherein:

When the described volume management information of record, described recording step is provided with the sector number of described predetermined quantity in described partition table.

17. the recording medium of claim 15, wherein:

Described file allocation table have a plurality of entrys, with some bunches of corresponding some entrys, and every entry has been represented in identical file the link to other bunch;

Described program may further comprise the steps:

Receiving step is used to be received in the described memory block setting to the sector sum;

Described calculation procedure comprises:

First calculation procedure, by with described sector sum divided by sector number 2 ^jCalculate total number of clusters, and calculate the size of described file allocation table by the bit length that described total number of clusters be multiply by an entry, and

Described reservation step comprises:

Second calculation procedure, by the size of described file allocation table and the sector number addition that constitutes described partition table and root directory entry respectively being calculated itself and SUM, and should { formula 1} calculates the m value with the SUM substitution.

18. the recording medium of claim 17, wherein:

Described memory block comprises: have only the protected location that could conduct interviews by this equipment when the equipment that links to each other with described semiconductor memory card authorized when checking and

No matter whether the mandate of described associated devices is verified the equal user data area that can be conducted interviews by this equipment;

Receive setting the source of described receiving step outside described access means to the sector number of distributing to each described protected location and user data area;

Described first calculation procedure is according to the size that is set at each described protected location and user data area calculation document allocation table that is received; And

Described second calculation procedure: (1) is respectively by each in the protected location is big or small and the user data area size and the sector number addition that constitutes the sector number of described subregion boot section information and constitute the root directory entry; so that calculate two and SUM of this protected location and user data area respectively; (2) with described two and in each SUM value substitution { formula 1} calculates the m value of this protected location and user data area respectively.

19. the initial method of an initializing computer is used for by with every group 2 ^jData on (j be 0 or positive integer) individual sector manage as one bunch, and one or more bunches managed as a file and execute file visit on semiconductor memory card, described semiconductor memory card has one and forms the memory block by a plurality of sectors, in this memory block every group 2 ⁱIndividual contiguous sector is formed a piece (i be 0 or positive integer), and piece is the least unit that can carry out data erase, and this initial method may further comprise the steps:

Calculation procedure calculates the size of volume management information according to be processed bunch quantity in the described memory block, and described volume management information comprises a file allocation table, is used to each file indication and this document to linking between each bunch accordingly;

Reserve step, reserve (1) first area that is used to write down volume management information, and (2) second areas that are used for user data, the data length that described first area had is greater than the volume management information of being calculated, and by m * 2 ^j(m is a positive integer) individual sector is formed, and described second area is made up of the sector of following the first area; And

Recording step, record volume management information in the first area, user data in second area, and with these volume management information and user data as bunch managing.

20. a semiconductor memory card comprises a memory block of being made up of a plurality of sectors, and with every group 2 ^jData on (j be 0 or positive integer) individual sector manage as one bunch, and one or more bunches are managed as a file, in this memory block every group 2 ⁱIndividual contiguous sector is formed a piece (i be 0 or positive integer), and piece is the least unit that can carry out data erase, and this semiconductor memory card comprises:

The first area is by m * 2 ^j(m is a positive integer) individual sector is formed, and is used to write down volume management information; And

Second area is made up of the sector of following the first area, is used for user data;

Described volume management information comprises a file allocation table, is used to the indication of each file and this document to linking between each bunch accordingly.

21. the semiconductor memory card of claim 20, wherein:

Except described file allocation table, described volume management information also comprises Main Boot Record, partition table, subregion boot section information, root directory entry; And

Described Main Boot Record of record and partition table in first sector of described first area, after skipping the sector of predetermined quantity, the described subregion of record boot section information, file allocation table and root directory entry in follow-up sector are so that make the end of described first area consistent with the end of root directory entry.

22. the semiconductor memory card of claim 20 has also write down the attribute list that comprises many attribute entrys, every entry is corresponding to one of some;

The user data that writes down in described second area has as at least one to be write the file of forbidding attribute and manages; And

Described record cell can be operated to be used for writing for the following entry setting in the described attribute list forbids attribute, promptly described entry corresponding to fully by with write the piece that the corresponding sector of each bunch of forbidding in the file is constituted.

23. semiconductor memory card that contains the memory block; this memory block comprises: (1) protected location; can only conduct interviews by the equipment that links to each other with described semiconductor memory card and its mandate has been verified; and (2) user data area; no matter whether the mandate of described associated devices is verified; all can be conducted interviews by this equipment, this protected location and user data area include a plurality of sectors, and with every group 2 ^jData on (j be 0 or positive integer) individual sector manage as one bunch, and one or more bunches are managed every group 2 as a file ⁱIndividual contiguous sector is formed a piece (i be 0 or positive integer), and piece is the least unit that can carry out data erase, and in described protected location and the user data area at least one comprises:

The first area is by m * 2 ^j(m is a positive integer) individual sector is formed, and is used to write down volume management information; And

Second area is made up of the sector of following the first area, is used for user data;

Described volume management information comprises a file allocation table, is used to the indication of each file and this document to linking between each bunch accordingly.

24. the semiconductor memory card of claim 23, wherein:

Except described file allocation table, described volume management information also comprises Main Boot Record, partition table, subregion boot section information and root directory entry; And

Described Main Boot Record of record and partition table in first sector of described first area, after skipping the sector of predetermined quantity, the described subregion of record boot section information, file allocation table and root directory entry in follow-up sector are so that make the end of described first area consistent with the end of root directory entry.

25. the semiconductor memory card of claim 23 has also write down the attribute list that comprises many attribute entrys, every entry is corresponding to one of some;

The user data that writes down in described second area has as at least one to be write the file of forbidding attribute and manages; And

Described record cell can be operated to be used for writing for the following entry setting in the described attribute list forbids attribute, promptly described entry corresponding to fully by with write the piece that the corresponding sector of each bunch of forbidding in the file is constituted.

Images